Human posture recognition for estimation of human body condition

Wei Quan, Jinseok Woo, Yuichiro Toda, Naoyuki Kubota

Research output: Contribution to journalArticle

Abstract

Human posture recognition has been a popular research topic since the development of the referent fields of human-robot interaction, and simulation operation. Most of these methods are based on supervised learning, and a large amount of training information is required to conduct an ideal assessment. In this study, we propose a solution to this by applying a number of unsupervised learning algorithms based on the forward kinematics model of the human skeleton. Next, we optimize the proposed method by integrating particle swarm optimization (PSO) for optimization. The advantage of the proposed method is no pre-training data is that required for human posture generation and recognition. We validate the method by conducting a series of experiments with human subjects.

Original languageEnglish
Pages (from-to)519-527
Number of pages9
JournalJournal of Advanced Computational Intelligence and Intelligent Informatics
Volume23
Issue number3
DOIs
Publication statusPublished - May 1 2019

Fingerprint

Human robot interaction
Unsupervised learning
Supervised learning
Particle swarm optimization (PSO)
Learning algorithms
Kinematics
Experiments

Keywords

  • Growing neural gas
  • Human posture recognition
  • Human-robot interaction
  • Particle swarm optimization

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Human posture recognition for estimation of human body condition. / Quan, Wei; Woo, Jinseok; Toda, Yuichiro; Kubota, Naoyuki.

In: Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol. 23, No. 3, 01.05.2019, p. 519-527.

Research output: Contribution to journalArticle

@article{f526c14b32564c25858f3b625983d29b,
title = "Human posture recognition for estimation of human body condition",
abstract = "Human posture recognition has been a popular research topic since the development of the referent fields of human-robot interaction, and simulation operation. Most of these methods are based on supervised learning, and a large amount of training information is required to conduct an ideal assessment. In this study, we propose a solution to this by applying a number of unsupervised learning algorithms based on the forward kinematics model of the human skeleton. Next, we optimize the proposed method by integrating particle swarm optimization (PSO) for optimization. The advantage of the proposed method is no pre-training data is that required for human posture generation and recognition. We validate the method by conducting a series of experiments with human subjects.",
keywords = "Growing neural gas, Human posture recognition, Human-robot interaction, Particle swarm optimization",
author = "Wei Quan and Jinseok Woo and Yuichiro Toda and Naoyuki Kubota",
year = "2019",
month = "5",
day = "1",
doi = "10.20965/jaciii.2019.p0519",
language = "English",
volume = "23",
pages = "519--527",
journal = "Journal of Advanced Computational Intelligence and Intelligent Informatics",
issn = "1343-0130",
publisher = "Fuji Technology Press",
number = "3",

}

TY - JOUR

T1 - Human posture recognition for estimation of human body condition

AU - Quan, Wei

AU - Woo, Jinseok

AU - Toda, Yuichiro

AU - Kubota, Naoyuki

PY - 2019/5/1

Y1 - 2019/5/1

N2 - Human posture recognition has been a popular research topic since the development of the referent fields of human-robot interaction, and simulation operation. Most of these methods are based on supervised learning, and a large amount of training information is required to conduct an ideal assessment. In this study, we propose a solution to this by applying a number of unsupervised learning algorithms based on the forward kinematics model of the human skeleton. Next, we optimize the proposed method by integrating particle swarm optimization (PSO) for optimization. The advantage of the proposed method is no pre-training data is that required for human posture generation and recognition. We validate the method by conducting a series of experiments with human subjects.

AB - Human posture recognition has been a popular research topic since the development of the referent fields of human-robot interaction, and simulation operation. Most of these methods are based on supervised learning, and a large amount of training information is required to conduct an ideal assessment. In this study, we propose a solution to this by applying a number of unsupervised learning algorithms based on the forward kinematics model of the human skeleton. Next, we optimize the proposed method by integrating particle swarm optimization (PSO) for optimization. The advantage of the proposed method is no pre-training data is that required for human posture generation and recognition. We validate the method by conducting a series of experiments with human subjects.

KW - Growing neural gas

KW - Human posture recognition

KW - Human-robot interaction

KW - Particle swarm optimization

UR - http://www.scopus.com/inward/record.url?scp=85068001849&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068001849&partnerID=8YFLogxK

U2 - 10.20965/jaciii.2019.p0519

DO - 10.20965/jaciii.2019.p0519

M3 - Article

VL - 23

SP - 519

EP - 527

JO - Journal of Advanced Computational Intelligence and Intelligent Informatics

JF - Journal of Advanced Computational Intelligence and Intelligent Informatics

SN - 1343-0130

IS - 3

ER -